Depositional controls on glacial facies associations in a basinal setting, late proterozoic, amadeus basin, Central Australia

Abstract

Diamictites, many established as being glacial in origin, are globally distributed in Late Protezoic rocks. Their widespread occurrence is important in that it implies that significant climatic events occurred during the period between 900 and 600 Ma. However, because of the age of the rocks, correlations are poor and it is difficult to establish, in a convincing manner, whether these sediments were deposited synchronously, during a major global ice age, or if each occurrence of diamictite relates to local climatic or tectonic events. Since the biostratigraphy of these rocks is limited and only a small number of radiometric ages are available, an alternative approach to correlation is necessary. The most promising approach currently available appears to be the application of sequence stratigraphy and the careful evaluation of facies associations on a site by site basis before attempting global correlations. Late Proterozoic diamictites occur in a number of intracratonic basins in a broad belt across the ancient Australian craton. Two Late Proterozoic diamictite units, both previously identified as glacial in origin, are exposed in the intracratonic Amadeus basin of central Australia. The Areyonga Formation, the older of the two units, forms the basis of this study. The Areyonga Formation is contained within a major depositional sequence, the Areyonga sequence, the bounds of which were determined by basin dynamics and eustatic sea-level changes induced by an ice age of global dimensions. During deposition of this sequence three major sedimentological systems were in operation. Along the northern margin of the basin glaciation dominated sedimentation. In the south large volumes of sediment were deposited in a shallow-marine setting whilst to the west braided streams supplied sediment to a major coastal plain and delta system. All three systems operated within the same sea-level and basinal constraints but each contributed sediment to the basin independently. The glacial sediments of the Areyonga sequence were deposited from a relatively small ice cap lying between the Amadeus Basin and the Ngalia Basin to the north. Sediments were deposited from two tongues of ice that extended about 100 km into the northeastern corner of the Amadeus Basin. Glacial sedimentation occurred in three stages determined by the initial fall and subsequent rise in eustatic sea level as the ice sheet grew and declined. Initially thick massive basal till units were deposited on an erosion surface intersected by subglacial and ice-marginal channels. They were then followed by thinner till units separated by channel-like sandstone and conglomerate with associated boulder surfaces or concentrations that were deposited by an extensive subglacial drainage network. Finally, the subglacial deposits grade into ice-proximal and ice-distal shallow-marine sediments deposited in part by mass movement. Sedimentation appears to have been terminated by a major sea-level fall. In the Ooraminna sub-basin to the east, rocks, originally thought to have been deposited during a second glaciation, appear to be part of a lowstand wedge formed during a sea-level low following the Areyonga glaciation. A record of the second glaciation probably does not exist in the Amadeus basin. Similarly, diamictites in the Inindia beds and originally correlated with the Areyonga Formation have been shown to belong to the non-glacial lowstand wedge of the following sequence. Preservation of glacial sediments is most likely to occur in a basinal setting where depositional space is being actively created. In the Amadeus basin depositional space was created prior to deposition of the Areyonga Formation by subsidence following crustal extension that occurred approximately 150 m.y. previously. The distribution of sequence boundaries and the succession of facies associations identified in the Areyonga sequence provide the basis for a depositional model for glacial sedimentation during an ice age. To be successful, correlation of Late Proterozoic glacial sediments must be based on an understanding of these facies associations and their bounding depositional surfaces. Associations similar to those found in the Amadeus basin have been documented in other Proterozoic basins suggesting that the facies associations and relationships recorded in the Areyonga sequence may form the basis for a generalized model for basinal sedimentation from an ice age of global significance. If this is so, it should be possible to determine if a sequence containing glacial sediments is part of a global climatic event or simply due to local conditions from a single site evaluation.